Feasibility Studies on Si-Based Biosensors - MDPI

3 downloads 44 Views 800KB Size Report
May 11, 2009 - Sebania Libertino 1,*, Venera Aiello 2,3, Antonino Scandurra 4, ..... electrically characterized and the results are compared in Figure 13.
Sensors 2009, 9, 3469-3490; doi:10.3390/s90503469 OPEN ACCESS

sensors

ISSN 1424-8220 www.mdpi.com/journal/sensors Review

Feasibility Studies on Si-Based Biosensors Sebania Libertino 1,*, Venera Aiello 2,3, Antonino Scandurra 4, Marcella Renis 2, Fulvia Sinatra 3 and Salvatore Lombardo 1 1 2

3

4

CNR – IMM Catania, Italy; E-Mail: [email protected] Università degli Studi di Catania, Dipartimento di Chimica Biologica, Chimica Medica e Biologia Molecolare, Catania, Italy; E-Mails: [email protected]; [email protected] Università degli Studi di Catania, Dipartimento di Scienze Biomediche, Catania, Italy; E-Mail: [email protected] Laboratorio Superfici e Interfasi (SUPERLAB), Consorzio Catania Ricerche, Catania, Italy; E-Mail: [email protected]

* Author to whom correspondence should be addressed; E-Mail: [email protected]; Tel.: (+39)0955968224. Fax: (+39)0955968312 Received: 20 February 2009; in revised form: 6 April 2009 / Accepted: 9 April 2009 / Published: 11 May 2009

Abstract: The aim of this paper is to summarize the efforts carried out so far in the fabrication of Si-based biosensors by a team of researchers in Catania, Italy. This work was born as a collaboration between the Catania section of the Microelectronic and Microsystem Institute (IMM) of the CNR, the Surfaces and Interfaces laboratory (SUPERLAB) of the Consorzio Catania Ricerche and two departments at the University of Catania: the Biomedical Science and the Biological Chemistry and Molecular Biology Departments. The first goal of our study was the definition and optimization of an immobilization protocol capable of bonding the biological sensing element on a Si-based surface via covalent chemical bonds. We chose SiO2 as the anchoring surface due to its biocompatibility and extensive presence in microelectronic devices. The immobilization protocol was tested and optimized, introducing a new step, oxide activation, using techniques compatible with microelectronic processing. The importance of the added step is described by the experimental results. We also tested different biological molecule concentrations in the immobilization solutions and the effects on the immobilized layer. Finally a MOS-like structure was designed and fabricated to test an electrical transduction

Sensors 2009, 9

3470

mechanism. The results obtained so far and the possible evolution of the research field are described in this review paper. Keywords: Si-based biosensors; biological molecules immobilization; glutaraldehyde; glucose oxidase; DNA strands; metallothioneines; microelectronic compatibility

1. Introduction The ability to detect biomolecular interactions is of extreme importance in medical, pharmaceutical and biotechnological research and development. Biosensors have been developed for this purpose [13]. Their increasing importance in everyday life is driving a merger of the microelectronics and biomedical communities. The common objective is the production of devices ready for mass production that will perform accurate analyses. A biosensor is a device that transforms biochemical information (presence and/or concentration of a specific analyte), into an analytically useful signal. It can be schematically represented as two basic components connected in series: a biological recognition system (bio-receptor, usually acting with interactions at supramolecular level) and a physical-chemical transducer. The system may be completed by a signal amplifier and a microelectronic circuit to elaborate the signal. Usually, the biosensor and the signal processing circuitry are not integrated. Different types of biologically sensitive materials can be applied as recognition elements. They can be enzymes, antibodies, antigens [4], proteins [5], nucleic acids [6,7] or even living biological systems (e.g., cells, plants, organs or whole organisms) [8]. Over the last 20 years there has been a growing interest in creating microbiosensors, fabricated in Si-compatible technologies, to be integrated within microelectronic circuits. The reason is that siliconbased devices would provide a lot of potential advantages such as small size and weight, fast response, high reliability, low output impedance, the possibility of automatic packaging at wafer level, on-chip integration and a signal processing scheme with the future prospect of low-cost mass production of portable microanalysis systems. In fact, among microelectronic materials, silicon (Si) has the most mature and low cost technology. Moreover, the Metal-Oxide-Semiconductor (MOS) system based on silicon as semiconductor and on SiO2 as dielectric is one of the key enabling technologies of these last fifty years. Complementary MOS (CMOS) technology has allowed the development of VLSI circuits with unprecedented performances at an exceptionally low cost for most of digital, analog, mixed signal, and RF circuits. The basis for this impressive progress is the exceptional quality of the Si/SiO2 system in terms of interface and bulk defects, low cost robustness to electrical and mechanical stress, scalability of the transistors, etc. Finally, SiO2 based matrixes have been proved to be a very useful support for the immobilization of biological molecules thanks to their capability of retaining biological activity. Many goals will be achieved using Si-based materials: i) the possibility to shrink the devices, implying reduced molecular diffusion path, faster kinetics and an improvement of the analytical performance of the device [9,10]; ii) the possibility to create micro-structured devices achieving complex functions, e.g. micro-total-analysis-systems; iii) the integration on the same chip of the electronics and/or photo-electronics needed for detection; iv) the possibility to make in vivo

Sensors 2009, 9

3471

physiological monitoring; it implies lower reagent consumption, hence minimized sample volumes, lower energy consumption, and less space requirement (sensor portability). It should be mentioned that conventional biosensors need extensive packaging, complex electronic interfaces and regular maintenance or reactivation. Finally, electrical sensing is considered one of the main goals to achieve in the next generation of biosensors since it could promote their integration within complex electrical circuits. Due to their simple principle of measurement and integrable signal processing on chip, biosensors that are based on electrochemical transducer principles are the most common sensor devices produced so far [11-20] and used in everyday life. Moreover, they are the most easily integrable in a microelectronic circuit, thus they would minimize the integration efforts in a complex circuitry. The immobilization of the biological probe onto the transducer surface plays an important role in the overall performances of biosensors. Three main issues must be considered: i) the sensing surface must be biocompatible; ii) the immobilization protocol must not degrade the inorganic part of the sensor (for Si-based devices, VLSI compatibility); iii) the biological molecule must be anchored to the solid surface avoiding its denaturizing or the loss of its activity. The environment of the immobilized probes at the solid surface depends upon the mode of immobilization and can differ from that experienced in the bulk solution. These issues have been the target of many works in literature [10,2126]. The most used approach is the formation of covalent bonds with the solid surface [10,24,25,2729], often using bifunctional reagents to bridge the biological molecule and the functionalized sample surface. If silicon dioxide is used, the thermal processing used (wet or dry) may produce quite different results [23]. The immobilization procedure must be optimized to obtain the maximum surface coverage and to prevent the biological molecule denaturation and/or the loss of its specific property, e.g. for an enzyme its enzymatic activity [30,31]. Si-based biosensors, as well as conventional microelectronic devices, must be fully characterized using standard microelectronics techniques allowing biological molecule monitoring. In this way, the new technology costs are contained, since no new equipment is needed. Different techniques were used: X-ray Photoelectron Spectroscopy (XPS) and Atomic Force Microscopy (AFM). The first one provides information on chemical bonds and molecular composition of the material surfaces, combined with a high surface specificity; while the second one allows a careful topographic inspection of the surface. Finally, spectrophotometric techniques were also used. The above mentioned techniques were used to study the immobilization of three different biological molecules on SiO2 surfaces: glucose oxidase (GOx), DNA strands and metallothioneine (MT). GOx is 160 kDa homodimeric globular protein, with a tightly bound (Ka = 1x10-10) flavin adenine dinucleotide (FAD) per monomer. The overall dimer dimensions measured by X-ray crystallography are 6.0×5.2×7.7 nm3 [32]. GOx, as all peptides and proteins, is a polymer of α-amino acids; it includes 580 amino acids, the FAD cofactor, six N-acetylglucosamine residues, three mannose residues and 152 solvent molecule acids [32]. As it is well known, the general chemical structure of an α-amino acid (excluding proline) is R-CH-NH2–(COOH). The enzyme catalyzes the oxidation of β-D-glucose to D-glucono-1,5-lactone by a reaction that can be summarized in two steps: i) glucose oxidation with the enzyme reduction, ii) re-oxidation of the enzyme with consumption of molecular oxide (O2) and production of hydrogen peroxide (H2O2) [33]. This enzyme is usually employed when the glucose

Sensors 2009, 9

3472

concentration in the blood must be measured, hence GOx based micro-biosensors [34] would have immediate applications in monitoring diabetes [35]. The second biological molecule immobilized was single strand DNA (ssDNA). DNA molecules are charged macromolecules and the direct hybridisation event is an affinity binding process. DNA is a poly-anion with negative charges along its phosphate backbone. Double strand DNA can be considered as a circular cylinder (with a diameter of about 1.5–2 nm) with electrostatic charges evenly distributed about the cylindrical surface [36]; the length of a DNA probe depends on the number of nucleotides and the length of a nucleotide (or base) is about ~ 0.34 nm [6]. To selectively recognise a unique human DNA sequence, DNA probes must be at least 16 bases long [37]. DNA recognition methods have assumed a primary importance in the genetic diseases’ diagnosis. Finally, MTs, extracted from rabbit liver, have a low molecular weight (